System for evaluating hippocampal function

Abstract

In the present invention, a hippocampal function is evaluated, while reflecting the effect of a physical exercise using a motion function improving device in a subject, in the state where blood flow is improved by dilating peripheral blood vessels around the head of the subject, and at the same time, synaptic strengthening is effectively induced.

Description

Hippocampal function evaluation system

[0001] The present invention proposes a hippocampal function evaluation system for evaluating hippocampal function in the brain of a subject.

[0002] As Japan's population ages, the number of dementia patients in the country is on the rise. In particular, vascular dementia and Alzheimer's disease account for the majority of dementia cases, and many patients exhibit symptoms of both diseases.

[0003] Common methods of treating vascular dementia include drug treatment for high blood pressure, dyslipidemia, and diabetes, as well as promoting exercise, quitting smoking, preventing overeating, and reducing stress. However, no pharmaceutical treatment for vascular dementia itself has yet been developed.

[0004] Furthermore, in Alzheimer's disease, degeneration of nerve cells is observed more than 20 years before the onset of the disease, and it is believed that this leads to the loss of nerve cells, resulting in atrophy of the hippocampus and a decline in hippocampal function. Therefore, while it may be possible to delay or prevent the onset of Alzheimer's disease by taking measures to improve or maintain hippocampal function, a fundamental treatment has not yet been developed.

[0005] In recent years, there have been many reported cases in which subjects' functions have improved by using a motor function improvement device that can control and assist movement based on the bioelectric potential associated with voluntary muscle activity in accordance with the subject's intention when performing treatment or rehabilitation on subjects with functional disorders of the nervous system or physical system.

[0006] Based on signals from the subject's nervous system, the motor function improvement device functions to move the subject's body, which has impaired motor function. As a result, the subject's musculoskeletal system moves at will, and information from the sensory system flows to the nervous system through both inside and outside the body, creating a two-way biofeedback loop between the nervous system and the musculoskeletal system.

[0007] It is believed that repeating this process strengthens synaptic connections in the brain, nerves, and muscles, promoting relearning and functional regeneration, and facilitating improvement of physical functions in subjects with brain, nervous, or muscular disorders (see, for example, Patent Document 1).

[0008] Furthermore, a method for calculating an evaluation value of hippocampal function, which serves as an index for determining the health of a subject's hippocampal function, has been proposed (see, for example, Patent Document 2). This method involves having the subject perform a behavioral analysis task, and determining the state of newborn neurons in the hippocampus involved in the hippocampal pattern separation ability (the ability to separate and differentiate similar memories or events) based on the repeated trials, responses, and correct answer rate.

[0009] JP 2024-65996 A Japanese Patent No. 6328469

[0010] When evaluating a subject's hippocampal function using the method described in Patent Document 2, if the motor function improvement device described in Patent Document 1 is applied, it is predicted that the evaluation result will include an improvement in the subject's physical functions, particularly in hippocampal function.

[0011] It is also expected that the results of the evaluation of the subjects' hippocampal function will contribute to providing useful information for delaying or preventing the onset of Alzheimer's disease.

[0012] The present invention has been made in consideration of the above points, and aims to propose a hippocampal function evaluation system that can accurately evaluate the improvement state of a subject's hippocampal function while reflecting the effects of the subject's physical exercise using a motor function improvement device.

[0013] In order to solve this problem, the present invention provides a movement function improvement device that promotes the effect of improving the motor function of the subject's brain, nerves, and muscles by implementing a bidirectional biofeedback loop between the brain, nervous system, and musculoskeletal system, which is constructed while correcting the difference between the movement commands from the brain, nervous system, and actual movement phenomena through the repetition of the subject's voluntary physical movements; an information acquisition device that acquires information related to the subject's hippocampal function during or after the movement of the movement function improvement device while the movement function improvement device is worn by the subject; and an evaluation device that evaluates the subject's hippocampal function based on the information acquired by the information acquisition device.

[0014] In this way, the hippocampal function evaluation system reflects the effects of the subject's physical exercise using the motor function improvement device, and by evaluating hippocampal function while dilating peripheral blood vessels, mainly around the subject's head, to improve blood flow and effectively induce synaptic strengthening, it becomes possible to accurately recognize whether the subject's hippocampal function has been improved.

[0015] In addition, in the present invention, the information includes at least one of information regarding hippocampal neogenetic ability, synaptic function, pattern separation ability, pattern completion ability, formation of memory traces related to motor procedures in the cerebellum, motor-related functional areas, and somatosensory areas.

[0016] As a result, the hippocampal function evaluation system, by comprehensively including information related to the subject's hippocampal function, can contribute to methods for improving trunk function that are less stressful on the body (memory trace formation methods, synapse strengthening methods) and methods for improving hippocampal function (methods for improving cognitive function, methods for maintaining cognitive function, methods for preventing cognitive decline).

[0017] Furthermore, in the present invention, the evaluation device sequentially analyzes the form of the information acquired by the information acquisition device, extracts feature data corresponding to each of the forms, and then uses feature patterns classified according to indicators for determining the health of hippocampal function in healthy individuals as training data, and evaluates hippocampal function according to the corresponding indicators from the sequentially extracted feature data while referring to a health model constructed by deep learning.

[0018] As a result, when the hippocampal function evaluation system evaluates the hippocampal function of a subject, it is possible to significantly improve the accuracy of evaluation of the hippocampal function according to the index.

[0019] Furthermore, in the present invention, a motor function improving device is used to perform voluntary physical movements of the subject, which includes a drive unit that applies power to the subject, a signal detection unit that detects the subject's biopotential signals, a biosignal processing unit that acquires the subject's nerve conduction signals and myoelectric potential signals from the biopotential signals detected by the signal detection unit, a voluntary control unit that uses the nerve conduction signals and myoelectric potential signals acquired by the biosignal processing unit to generate command signals for causing the drive unit to generate power in accordance with the subject's will, and a drive current generation unit that generates currents corresponding to the nerve conduction signals and myoelectric potential signals based on the command signals generated by the voluntary control unit and supplies them to the drive unit.

[0020] In this way, when the subject repeatedly performs voluntary movements using the motor function improvement device, the motor function of the subject's brain, nerves, and muscles can be improved by correcting the difference between the movement commands from the nervous system and the actual motor phenomenon, which can effectively contribute to improving the subject's hippocampal function in the hippocampal function evaluation system.

[0021] Furthermore, the present invention includes a motion mechanism unit that is used to be integrated with the subject and has a drive unit that is driven actively or passively in conjunction with the subject's physical motion; a signal detection unit that detects changes in ionic current transmitted from the subject's nervous system to the muscular system as bioelectric potential signals that appear on the skin surface; a joint circumference detection unit that detects physical quantities around the joints that accompany the subject's physical motion based on the output signal from the drive unit; an optional control unit that controls the drive unit so as to produce a movement phenomenon that reflects the subject's intention to move based on the bioelectric potential signals and the physical quantities around the joints; and a data storage unit that stores reference parameters for each of the phases, which are a series of minimum motion units that make up the subject's motion pattern classified as a task, and that estimates the phase of the subject's task by comparing the physical quantities around the joints with the reference parameters stored in the data storage unit and applies power according to the phase. The subject's voluntary physical movements are performed using a movement function improvement device equipped with an autonomous control unit that controls the drive unit to generate a voluntary movement, and a synthesis control unit that stores the control ratios of the voluntary control unit and the autonomous control unit set for each phase of each task in a data storage unit and synthesizes the control states of the voluntary control unit and the autonomous control unit so as to achieve a control ratio corresponding to the phase.The synthesis control unit compensates for the physical impedance of the entire system consisting of the entire device and the subject, based on the physical quantities around the joints, in accordance with the physical characteristics of the entire system including the subject's body characteristics and gravity, and also feedback-adjusts the synthesized control state while correcting the difference using movement commands from the nervous system based on the biological self-control loop that is interactively promoted between the subject's body and the movement mechanism unit, so as to minimize the difference between the subject's movement intention and the motor phenomenon.

[0022] In this way, when the subject repeatedly performs voluntary movements using the movement mechanism in the motor function improvement device, the motor function of the subject's brain, nerves, and muscles can be improved by correcting the difference between the movement commands from the nervous system and the actual motor phenomenon, which can effectively contribute to improving the subject's hippocampal function in the hippocampal function evaluation system.

[0023] Furthermore, in the present invention, when forming a biological self-control loop, the smallest motor control unit for realizing voluntary movements caused by the subject's will is set as a minimum voluntary movement control unit consisting of the cranial nervous system, synaptic connections, and muscular system, and a movement mechanism is used to establish a biological self-control loop for each minimum voluntary movement control unit that forms linked physical movements to realize a movement phenomenon.

[0024] In this way, by explicitly incorporating the influence of the diseased area and disease cause on the minimum voluntary movement control unit in the movement function improvement device into the process of function improvement and treatment based on the basic theory of the biological self-control loop by the movement mechanism, the effect is also felt on the other minimum voluntary movement control units, and they work together in sync with the movement of the movement mechanism, and the function of each minimum voluntary movement control unit is strengthened and adjusted in sync with the movement of the movement mechanism to achieve the target movement, thereby improving the function of the brain, nerves, and muscular systems.

[0025] According to the present invention, a hippocampal function evaluation system can be realized that can accurately evaluate the improvement state of a subject's hippocampal function while reflecting the effects of the subject's physical exercise performed by a motor function improvement device.

[0026] 5 is a conceptual diagram illustrating the basic theory of a biological autoregulation loop according to the present invention. FIG. 6 is a conceptual diagram illustrating a minimum voluntary movement control unit. FIG. 7 is a conceptual diagram illustrating a transition state when the basic theory of a biological autoregulation loop described above is applied to a subject. FIG. 8 is a schematic diagram illustrating the external configuration of a movement function improving device including a waist-type movement mechanism unit according to the present embodiment. FIG. 9 is a schematic diagram illustrating the main components of the movement function improving device of FIG. 4. FIG. 10 is a schematic diagram illustrating the movement states and ranges of motion of the movement function improving device of FIG. 4. FIG. 11 is a block diagram illustrating the configuration of the control system of the movement function improving device of FIG. 4. FIG. 12 is a conceptual diagram illustrating an example of each task and each phase stored in a data storage unit. FIG. 13 is a diagram illustrating the effect of function improvement using the movement function improving device. FIG. 14 is a diagram illustrating the effect of function improvement using the movement function improving device. FIG. 15 is a block diagram illustrating one embodiment of a hippocampal function evaluation system according to the present invention. FIG. 16 is a flowchart of the process by which the hippocampal function evaluation system evaluates the hippocampal function of a subject. FIG. 17 is a schematic diagram illustrating the protocol for acquiring data before and after HAL training using HAL (both lower limbs type). FIG. 18 is a diagram showing changes in quadriceps muscle strength before and after HAL training using HAL (both lower limbs type). 1 shows the results of MRI image analysis before and after HAL training using the HAL both lower limbs type. The active area is evaluated by BOLD response. This figure shows the results of simultaneous measurement using fMRI and 256-channel high-density electroencephalography before and after HAL training using the HAL both lower limbs type. This figure shows the results of ERP and time-frequency analysis of a movement (flexion and extension) imagery task before and after HAL training using the HAL both lower limbs type. This figure shows the correct answer rate for the Lure task before and after HAL training using the HAL lumbar type for independent living support. This figure shows the correct answer rate for each task in a hippocampal function evaluation test before training using the HAL lumbar type for independent living support, and time-frequency analysis. This figure shows the correct answer rate for each task in a hippocampal function evaluation test after training using the HAL lumbar type for independent living support, and time-frequency analysis. This figure is a schematic diagram used to explain an image conversion method using the sliding window method. This figure is a conceptual diagram showing a model structure for deep learning.

[0027] An embodiment of the present invention will be described in detail below with reference to the drawings.

[0028] (1) Basic theory of bio-self-control loop using a motor function improvement device As a method for improving the motor function of a subject's brain-nerve-muscle system, experimental research has been conducted medically to clarify information processing focusing on the brain. However, with brain research alone, it is difficult to target the motor function of the body because the brain is separated from the motor system.

[0029] The human motor system is composed of the central nervous system (brain and spinal cord), the peripheral nervous system, and the musculoskeletal system. When dealing with the flow of information from the efferent nerves that leave the center and head toward the periphery, these systems form a connected information transmission system, but this alone does not achieve appropriate motor control.

[0030] In other words, information from efferent nerves is transmitted to muscle fibers involved in muscle contraction, but information after contraction is transmitted from the brain via the spinal cord and motor nerves to muscle fibers, but this information does not return to the brain. For this reason, it is nearly impossible to clarify the mechanism for improving motor function or to develop techniques to improve motor control simply by analyzing the "motor unit" consisting of motor nerves and muscle fibers.

[0031] For this reason, some research has attempted to improve function by using robotic technology to move the joints and muscles of the legs and hands through external motion input, but it has also been reported that simply applying external force to move the joints and muscles of the legs and hands does not result in functional improvement.

[0032] In the present invention, a movement mechanism (the main mechanism of the movement function improvement device 10 shown in Figure 4, which will be described later) having a drive unit that drives actively or passively in conjunction with the subject's physical movements is used to demonstrate that when the subject performs a specific movement, as shown in Figure 1, command signals (efferent nerve signals) that cause the subject to move in accordance with their motor intention, and sensory signals (afferent nerve signals) generated by successfully moving, travel back and forth between the central system (brain and spinal cord) and peripheral system (motor nerves, muscular system, and sensory nerves), improving and reconstructing physical functions for voluntary movement.

[0033] The method for improving movement function according to the present invention comprises a voluntary control step for making the subject move in accordance with his or her movement intentions, an autonomous control step for generating a predetermined ideal power, and an impedance control step (including gravity compensation control) for reducing the feeling of difficulty in movement due to the load and viscous friction of the movement mechanism itself, thereby making it possible for the subject to feel as if the movement mechanism is a part of his or her own body, and achieving functional fusion and integration between the subject and the movement mechanism.

[0034] Furthermore, in this method for improving movement function, the synthesis control step that synthesizes the control states of the voluntary control step and the autonomous control step so as to achieve a control ratio according to the phase of the task not only executes the impedance control step described above, but also feedback-adjusts the synthesized control state while correcting the difference by movement commands from the nervous system based on a biological self-control loop that is interactively promoted between the subject's body and the movement mechanism, so as to minimize the difference between the subject's movement intention and the movement phenomenon.

[0035] This biological self-regulation loop is formed by the activation of proprioceptors called muscle spindles and tendon spindles in muscles and tendons in synchronization with the transmission of nervous system command information from the brain through the spinal cord and motor nerves to muscle fibers, causing muscle contraction. This activation information becomes information on muscle contraction and is fed back to the central nervous system (spinal cord and brain) via sensory nerves, and this feedback information is used to strengthen and adjust the synaptic connections between nerves and between nerves and muscles, in a repeated circular process.

[0036] In this way, in the method for improving motor function, when the subject repeatedly performs voluntary movements using the movement mechanism, the motor function of the subject's brain-nerve-muscle system can be improved by correcting the difference between the movement command from the brain-nerve system and the actual movement phenomenon.

[0037] One of the features of the present invention is that when forming a biological autoregulation loop, a minimum voluntary movement control unit consisting of the cranial nervous system (brain and spinal cord), synaptic connections (synaptic connections between nerves and between nerves and muscles), and the muscular system (muscle fibers (extrafusal muscle fibers connected to alpha motor neurons and intrafusal muscle fibers connected to gamma motor neurons), tendon fibers, muscle spindles, tendon spindles, etc.) can be configured as the minimum movement control unit for realizing voluntary movements caused by human will to establish a biological autoregulation loop.

[0038] In other words, the minimum voluntary movement control unit is a control unit for realizing the minimum voluntary movement formed by the pathway of the cranial nervous system (brain, spinal cord, motor nerves), muscle fibers, action (reaction), muscle spindles, tendon spindles, and the cranial nervous system (sensory nerves, spinal nerves, brain), as shown in Figure 2.

[0039] A minimum voluntary movement control unit that forms a specific joint movement is configured in multiple units, in conjunction with other minimum voluntary movement control units that are related to that minimum voluntary movement unit while linking with other joint movements, thereby achieving the desired overall movement.

[0040] In this process, the synaptic connections between nerves and between nerves and muscles in the minimum voluntary movement control units are adjusted and strengthened within the group of minimum voluntary movement control units related to the target overall movement and within the overall adjustment system, such as unconscious postural balance adjustment.

[0041] Furthermore, by using movement mechanisms (such as lower limb types, single-joint types, waist types, hand types, and finger types, which will be described later) for joint movements driven by muscle groups (so-called flexion and extension muscle groups) that are made up of the smallest voluntary movement control units (units), and for more complex coordinated movements made up of each joint system, it is possible to accommodate everything from the smallest units to higher-level complex body systems, thereby achieving functional improvements in the brain, nerves, and muscles.

[0042] In fact, if the minimum voluntary movement control unit is used as a unit to promote synaptic plasticity, neuroplasticity, and muscular plasticity, which are basic neural functions for improving the function of the brain, nerves, and muscles, the process for activating the subject's self-healing ability in response to the disease or symptoms (relaxation, stiffness, tremor, rigidity, ataxia, simultaneous contraction, etc.) will differ for each minimum voluntary movement control unit.

[0043] Therefore, the unit components of the minimum voluntary movement control unit for improving the motor function of the brain, nerves, and muscles differ for each disease or symptom, and the relevant parts and ranges of other minimum voluntary movement control units related to that minimum voluntary movement control unit also differ. Therefore, by using this as the minimum unit when making various adjustments to the movement mechanism section 20 (Figures 4 to 6 described below) in the movement function improvement device 10 according to the condition of the subject (such as tuning parameters to realize movement in accordance with the subject's movement intention and establish a biological self-control loop), it becomes possible to construct a treatment control strategy.

[0044] In the method for improving movement function according to the present invention, signals derived from the nervous system linked to voluntary will obtained from a pathway affected by the diseased area or cause of disease are detected, and focus is placed on the minimum voluntary movement control units involved in that pathway, and functional improvement is implemented for each minimum voluntary movement control unit that constitutes the overall voluntary movement that is the target.

[0045] The starting point for a human to perform voluntary movement is the expression of voluntary motor intent in the cerebrum. Nervous system signals of this motor intent are transmitted from the brain to the spinal cord, motor nerves, and muscle fibers, ultimately achieving the desired movement. However, because this process, from the expression of intent to the generation of movement, is too general, it is difficult to explicitly grasp the influence of disease sites and disease causes on voluntary movement, from the brain / nervous system through synaptic connections to the muscular system, including the sensory nerves of muscle fibers and muscle spindles and the gamma loop of gamma motor neurons. It is also difficult to explicitly implement treatments that take into account the influence of disease sites and disease causes. Therefore, in actual clinical settings, training is limited to the repetition of simple movements.

[0046] Therefore, by explicitly incorporating a minimum voluntary movement control unit that reflects the influence of the diseased area and disease cause into the process of functional improvement treatment based on the basic theory of the biological self-control loop using the movement mechanism unit 20 (Figures 4 to 6), the effect is also felt on the other minimum voluntary movement control units, and they work together in synchronization with the operation of the movement mechanism unit 20, and in order to achieve the target movement, each minimum voluntary movement control unit strengthens and adjusts the function of its unit components in synchronization with the operation of the movement mechanism unit, making it possible to improve the function of the brain, nerves, and muscles as a new method that differs from conventional methods.

[0047] 3A to 3E show transition states when the basic theory of the biological autoregulation loop described above is applied to a subject. Starting from a state in which the movement mechanism 20 is not attached before treatment (FIG. 3A), in the early stage of treatment, sensory nervous system information from the musculoskeletal system due to joint movement using the movement mechanism 20 is fed back to the central nervous system (brain and spinal cord) (FIG. 3B).

[0048] Subsequently, after the initial treatment, when the operating mechanism unit 20 is not attached (FIG. 3(C)), some feedback sensation of sensory nervous system information remains, but by performing continuous treatment using the operating mechanism unit 20 (FIG. 3(D)), the difference between the subject's intention to move and the motor phenomenon due to the operating command from the central nervous system is corrected based on the bio-autoregulation loop that is interactively promoted between the subject's body and the operating mechanism unit 20.

[0049] Even after continuous treatment, even when the operating mechanism 20 is not attached, the motor function of the brain, nerves, and muscles of the subject can be improved while activating the subject's self-healing power by correcting the difference between the operation command from the central nervous system and the actual motor phenomenon (Figure 3(E)).

[0050] (2) Configuration of the Motion Function Improving Device in the Present Embodiment (2-1) Configuration of the Motion Mechanism Unit (Waist Type: Hardware) Figures 4(A) and (B) show a motion function improving device 10 including a waist-type motion mechanism unit 20 in the present embodiment. Figures 5(A) and (B) show the main configuration of the motion mechanism unit 20 excluding the thigh cuff, belt, etc. The motion function improving device 10 is a device that assists the work and motion of a subject, and operates by detecting biopotential signals and the motion angle of the hip joint and the absolute angle of the trunk of the subject, and applying a driving force from a driving unit based on these detection signals.

[0051] When a subject wearing the device for improving movement function 10 voluntarily lifts and carries a relatively heavy object, the bioelectric potential signals generated on the skin surface of the latissimus dorsi or gluteus maximus (gluteus maximus) and the driving torque corresponding to the movement angle of the subject's hip joint are applied as an assisting force from the movement mechanism 20. Therefore, the subject can lift and carry the object using the combined force of his or her own muscle strength and the driving torque from the driving mechanism (actuator).

[0052] In addition to the carrying task of lifting an object and walking, the motor function improving device 10 can also assist in ascending and descending tasks, such as when the subject goes up and down stairs while carrying luggage.

[0053] In the movement mechanism unit 20 of the movement function improving device 10, a waist frame 30 is attached to the back side of the subject's waist in the left-right direction. The waist frame 30 is a hollow member made of, for example, CFRP (carbon fiber reinforced plastic), and has a rounded shape that matches the shape of the back and both sides of the waist of the human body.

[0054] The lumbar frame 30 is configured such that a first lumbar frame 30A is attached to the back side of the subject's lumbar region on both sides, and a second lumbar frame 30B is attached to the back side of the subject's lumbar region on both sides above the first lumbar frame 30A, connected via a support 31.

[0055] The first waist frame 30A and the second waist frame 30B are attached to the waist of the subject by attachment belts 32, 33 that are placed around the subject's abdominal side. When attached, the first waist frame 30A and the second waist frame 30B are attached in a forward-leaning position so that both ends are lower than the back side, i.e., so that both ends are located lower than the longitudinal center (the part located on the back side of the subject).

[0056] A left side frame 40 and a right side frame 41 are fixed to both ends of the first waist frame 30A and the second waist frame 30B.

[0057] In addition, on the outside opposite the side where the operating mechanism unit 20 is attached, a battery 42 is removably stored in the center of the first waist frame 30A, and wiring etc. connected to the battery 42 are inserted into the internal space.

[0058] The support pillar 31 is a member that vertically connects the longitudinal center of the first waist frame 30A and the longitudinal center of the second waist frame 30B. The support pillar 31 is a hollow member made of, for example, reinforced resin, through which sensor wiring is inserted. Also provided inside the support pillar 31 is a control device 80 (see FIG. 8, described below) that controls the operation of the operating mechanism 20.

[0059] The strength of the monocoque structure is ensured by supporting the various components of first waist frame 30A and second waist frame 30B so that they do not rotate, using support pillars 31. Attached to support pillars 31 are attachment belts 32 and 33, respectively, for first waist frame 30A and second waist frame 30B.

[0060] The left side frame 40 is fixed to the left of the subject's hip joint by joining the left end of the first waist frame 30A to the left end of the second waist frame 30B. The right side frame 41 has a structure that is approximately symmetrical to the left side frame 40, and is fixed by joining the right end of the first waist frame 30A to the right end of the second waist frame 30B.

[0061] An actuator and a brake mechanism (negative button 43) are built into the left frame 40, and are provided for inputting a minus (-) button for reducing the driving force of the actuator. The minus (-) button 43 is lit when the power of the motion function improving device 10 is on.

[0062] The right frame 41 houses an actuator and a brake mechanism (neither of which is shown), and is provided with a plus button 44 for inputting to increase the driving force of the actuator, and a power button 45 for switching on / off the power of the motion function improving device 10. The plus button 44 and the power button 45 are lit when the power of the motion function improving device 10 is on.

[0063] In this way, the waist frames (first waist frame 30A and second waist frame 30B), support pillars 41, and side frames (left side frame 40 and right side frame 41) are assembled into a single unit, thereby realizing a monocoque structure in which the frame itself bears stress.

[0064] 5A and 5B, the thigh fixing unit 50 comprises a left thigh fixing unit 50L for fixing the left thigh of the subject, and a right thigh fixing unit 50R for fixing the right thigh of the subject.

[0065] The left thigh fixing portion 50L is composed of a stay portion 51L connected to an actuator within the left side frame 40 and a belt portion 52L attached to the stay portion 51L, and is rotatable relative to the left side frame 40 in a side view.

[0066] The right thigh fixing portion 50R is composed of a stay portion 51R connected to an actuator inside the right side frame 41 and a belt portion 52R attached to the stay portion 51R, and is rotatable relative to the right side frame 41 in a side view.

[0067] In the left thigh fixing portion 50L and the right thigh fixing portion 50R, the stay portions 51L, 51R are designed to have an optimal length based on the average length of the thighs of the human body, and the subject's thighs are fixed by the belt portions 52L, 52R.

[0068] The attachment belt 32 is a attachment belt that is passed over the abdominal side when attaching the first waist frame 30A to the subject, and is the main attachment belt for attaching the operating mechanism unit 20 to the waist of the human body.

[0069] The attachment belt 33 is placed on the abdominal side when attaching the second waist frame 30B to the subject. The attachment belt 33 is used to fix the operation mechanism unit 20 to the human body above the attachment belt 33 in order to efficiently transmit the reaction force generated by the leg movement to the abdomen or waist of the subject when the subject wearing the operation mechanism unit 20 lifts a relatively heavy object with bent knees.

[0070] The battery 42 is located on the outside of the center of the first waist frame 30A, on the outside opposite to the side where the operating mechanism unit 20 is attached, and supplies power to the control device, actuator, brake mechanism (not shown), minus button 43, plus button 44 and power button 45.

[0071] The biosignal detection unit 60 having a biopotential sensor is attached to the back of the subject's waist and is a detection unit that detects biopotential signals associated with muscle activity when the subject tries to raise his or her trunk or when trying to maintain the angle of the trunk.

[0072] The three biopotential sensors of the biosignal detection unit 60 are connected to the ends of wires that extend from holes provided in the support 31 to the outside of the support 31. The biopotential sensors are attached to the back of the subject's waist to detect biopotential signals generated when the subject moves their trunk muscles.

[0073] The biopotential signals detected by the biosignal detection unit 60 are input to the control device. The biopotential sensors may be attached to the subject's back using tape or gel. One of the three sensors is used to measure the reference signal, and the remaining two sensors are used to measure the biopotential signals.

[0074] In practice, in the device 10 for improving movement function, the movement mechanism unit 20 is attached to the back side of the subject's waist. The device 10 is a device that generates an assist force to assist the movement of the subject's thighs relative to the waist when the subject stands up as shown in Fig. 6(B) from a bent position (squatting position) as shown in Fig. 6(A). Such movements of the subject include, for example, movements to stand up from a squatting position, movements to lift an object from a squatting position, movements during transfer assistance, etc.

[0075] Figure 6(C) is a diagram (left side view) showing the range of motion of the movement mechanism unit 20 in the movement function improving device 10. The left thigh fixing unit 50L can rotate 130° clockwise and 30° counterclockwise from the reference position shown in Figure 7(C) around the left side frame 40. Through this movement, the movement function improving device 10 generates an assist force to assist the movement of the thigh relative to the waist when the subject rises from a crouched position as shown in Figure 7(A) to a standing position as shown in Figure 7(B). The range of motion of the right thigh fixing unit 50R is similar.

[0076] The lower back-type motor function improving device 10 in this embodiment has a control system 70 as shown in Fig. 7, which will be described later. As a result, the motor function improving device 10 also has functionally a voluntary control step for making the subject move in accordance with their intention, an autonomous control step for generating a preset ideal power, and an impedance control step (including gravity compensation control) for reducing the feeling of difficulty in moving due to the load and viscous friction of the movement mechanism unit 20 itself.

[0077] As a result, with the waist-type movement function improvement device 10, the subject feels as if the movement mechanism unit 20 is a part of his or her body, making it possible to achieve functional fusion and integration between the subject and the movement mechanism unit 20.

[0078] In fact, with the motion function improving device 10, when the subject stands up from a bent position as shown in Fig. 6(A) to a crouched position as shown in Fig. 6(B), an assist force can be generated to assist the movement of the thighs relative to the waist. Such an assist force is generated by driving a drive unit (actuator) based on biopotential signals detected by the biosignal detection unit 60, which are associated with muscle activity when the subject tries to raise the trunk or muscle activity when trying to maintain the angle of the trunk.

[0079] Therefore, it is possible to provide a highly convenient motor function improving device 10 that can provide the necessary assistive force in the necessary direction according to the subject's will. It is also possible to provide a motor function improving device 10 that can minimize the power (muscle force) that the subject must generate himself or herself and prevent situations that impair the subject's convenience.

[0080] If the control device 80 (Figure 7) determines that the signal levels of the biological signals from the left and right thighs of the subject are not equal, it determines that the subject is walking, but if it determines that the signal levels are equal, it determines that the subject is stationary.

[0081] Furthermore, when the control device 80 determines that the subject is in a stationary state, if it determines that both the left and right hip joint angles are greater than a predetermined specified angle, it determines that the subject is walking, whereas if it determines that the angles are equal to or less than the specified angle, it determines that the subject is in a posture with the upper body lowered forward.

[0082] Furthermore, in the motion function improvement device 10, the synthesis control step that synthesizes the control states from the voluntary control step and the autonomous control step so as to achieve a control ratio according to the phase of the task not only executes the impedance control step described above, but also feedback-adjusts the synthesized control state while correcting the difference using motion commands from the nervous system based on a biological self-control loop that is interactively promoted between the subject's body and the motion mechanism, so as to minimize the difference between the subject's motion intention and the motor phenomenon.

[0083] As a result, in the motor function improvement device 10, when the subject repeatedly performs voluntary movements using the movement mechanism unit 20, the motor function of the subject's brain, nerves, and muscles can be improved by correcting the difference between the movement commands from the brain and nervous system and the actual motor phenomenon.

[0084] Furthermore, in the motor function improvement device 10, when forming a biological self-control loop, the smallest motor control unit for realizing voluntary movements caused by the subject's will is set as a minimum voluntary movement control unit consisting of the brain nervous system, synaptic connections, and muscular system, and a movement mechanism section 20 is used to establish a biological self-control loop for each minimum voluntary movement control unit that forms linked physical movements to realize a motor phenomenon.

[0085] As a result, in the movement function improvement device 10, by explicitly incorporating the influence of the diseased area and disease cause into the process of function improvement and treatment based on the basic theory of the biological self-control loop by the movement mechanism unit, the influence is extended to the other minimum voluntary movement control units, and they are linked in sync with the operation of the movement mechanism unit 20, and the function of each minimum voluntary movement control unit is strengthened and adjusted in sync with the operation of the movement mechanism unit to achieve the target movement, thereby improving the function of the brain, nerves, and muscular system.

[0086] (2-2) Control System in the Movement Function Improving Device Figure 7 is a block diagram showing the configuration of a control system 70 of the movement function improving device 10. As shown in Figure 7, the control system 70 of the movement function improving device 10 has a control device 80 that is responsible for overall control of the entire system, a data storage unit 81 in which various data is stored in a database that can be read and written in response to commands from the control device 80, and drive units 82L and 82R that are actively or passively driven in conjunction with the movement of the subject's lower limbs.

[0087] The control system 70 is also provided with a joint detection unit 90 having a potentiometer 83, an absolute angle sensor 84, and a torque sensor 85, and detects physical quantities around the joints associated with the subject's physical movements based on output signals from the driving units 82L and 82R.

[0088] The joint circumference detection unit 90 detects physical quantities around the joint, such as the absolute angle between the rotor side frame and the stator side frame of the drive units 82L, 82R in the operating mechanism unit 20, the rotation angle, angular velocity, angular acceleration, and drive torque.

[0089] The potentiometer 83 is coaxial with the output shaft of the actuator in the driving units 82L and 82R, and detects the rotation angle of the output shaft to detect the joint angle corresponding to the movement of the subject's lower limbs.

[0090] Absolute angle sensors 84 are mounted on the side frames (left frame 40 and right frame 41) and measure the absolute angle of the subject's thighs relative to the vertical direction. These absolute angle sensors 84 are composed of an acceleration sensor and a gyro sensor, and are used in sensor fusion, a method of extracting new information using data from multiple sensors.

[0091] To calculate the absolute angle of the thigh, a first-order filter is used to remove the effects of translational motion and temperature drift in each sensor. This first-order filter is calculated by weighting and adding the values ​​obtained from each sensor.

[0092] If the absolute angle of the thigh relative to the vertical direction is θabs(k), the angular velocity obtained by the gyro sensor is ω, the sampling period is dt, and the acceleration obtained by the acceleration sensor is α, then θabs(t) can be expressed as follows:

[0093] Furthermore, the torque sensor 85 detects the value of the current supplied to the driving units 82L and 82R, for example, and detects the driving torque by multiplying this current value by a torque constant specific to the actuator.

[0094] A biosignal detection unit 60 having a biopotential sensor (group of electrodes) is placed on the body surface of the subject (mainly the body surface of the thigh) based on the joints associated with the subject's lower limb movement, and detects changes in the ionic current transmitted from the subject's nervous system to the muscular system as a biopotential signal that appears on the skin surface.

[0095] The biosignal detection unit 60 is a detection unit that measures nerve action potentials emitted from the brain to the legs to move the subject's legs and muscle action potentials when skeletal muscles generate muscle force, and has electrodes that detect weak potentials generated in the periphery of the body system. In this embodiment, the biopotential sensor is attached so that it can be detachably attached to the surface of the subject's skin, for example, by an adhesive sticker that covers the periphery of the electrode.

[0096] The data storage unit 81 stores data required for various arithmetic processing in the control device 80. Biopotential signals detected by the biosignal detection unit 60 are stored in the data storage unit 81. Data on joint angles (θknee, θhip) detected by the absolute angle sensor 84 of the joint circumference detection unit 90 is input to the data storage unit 81.

[0097] The control device 80 is configured, for example, by a CPU (Central Processing Unit) chip having a memory, and includes a voluntary control unit 100, an autonomous control unit 101, and a synthesis control unit 102.

[0098] The optional control unit 100 controls the driving units 82L and 82R based on the bioelectric potential signal and the physical quantities around the joints so as to produce a movement phenomenon that reflects the subject's intention to move. Specifically, the optional control unit 100 supplies a command signal corresponding to the detection signal from the biosignal detection unit 60 to the power amplification unit 105.

[0099] The optional control unit 100 generates a command signal by applying a predetermined command function f(t) or gain P to the biological signal detection unit 60. This gain P is a preset value or function and can be adjusted by an external input.

[0100] Data on the knee joint angle detected by the potentiometer 83, data on the absolute angle of the thigh relative to the vertical direction detected by the absolute angle sensor 84, the driving torque detected by the torque sensor 85, and the biopotential signal detected by the biosignal detection unit 60 are input into the data storage unit 81.

[0101] The autonomous control unit 101 stores in the data storage unit 81 reference parameters for each phase, which is a series of smallest movement units that make up the movement pattern of the subject classified as a task, and estimates the phase of the subject's task by comparing the physical quantities around the joints with the reference parameters stored in the data storage unit 81, and controls the drive unit to generate power according to the phase.

[0102] The autonomous control unit 101 compares the knee joint angle data detected by the joint circumference detection unit (potentiometer 83) 90 with the knee joint angle of the reference parameter stored in the data storage unit 81, and estimates the phase of the subject's movement based on the comparison result.

[0103] Then, when the autonomous control unit 101 obtains the control data for the estimated phase, it generates a command signal according to the control data for this phase and supplies the command signal to the power amplifier unit 105 to cause the drive units 82L and 82R to generate this power.

[0104] Furthermore, the autonomous control unit 101 receives a gain adjusted by an external input, generates a command signal according to this gain, and outputs it to the power amplifier 105. The power amplifier 105 controls the current that drives the actuators of the drive units 82L and 82R to control the magnitude of the torque and the rotation angle of the actuators, thereby applying an assist force from the actuators to the knee joint of the subject.

[0105] In this way, the autonomous control unit 101 identifies the phases corresponding to the subject's task based on the physical quantities detected by the joint circumference detection unit (potentiometer 83, absolute angle sensor 84, and torque sensor 85) 90, and causes the drive units 82L and 82R to generate power corresponding to each phase.

[0106] The synthesis control unit 102 synthesizes the control signals from the voluntary control unit 100 and the autonomous control unit 101, and the drive current corresponding to the synthesized control signal is amplified by the power amplifier 105 and supplied to the actuators of the drive units 82L and 82R. The torque of this actuator is transmitted as an assist force to the knee joint of the subject via the waist frame 30.

[0107] 8 shows an example of each task and each phase stored in the data storage unit 81. As tasks for classifying the subject's movements, for example, task A having standing movement data for transitioning from a sitting position to a standing position, task B having walking movement data for the subject walking after standing, task C having sitting movement data for transitioning from a standing position to a sitting position, and task D having stair climbing movement data for ascending and descending stairs from a standing position are stored in the data storage unit 81.

[0108] Each task is set with multiple phase data. For example, task B for walking movement has phase B with movement data (joint angles, trajectory of center of gravity position, torque fluctuations, changes in bioelectric potential signals, etc.) when trying to swing the right leg forward from a standing position with the center of gravity on the left leg; phase B with movement data when landing from a position with the right leg forward and shifting the center of gravity; phase B with movement data when trying to swing the left leg forward from a standing position with the center of gravity on the right leg; and phase B with movement data when landing from a position with the left leg forward of the right leg and shifting the center of gravity.

[0109] In this way, by analyzing general human movements, it is found that there are fixed typical movement patterns, such as the angles of each joint and the movement of the center of gravity, in each phase. Therefore, for each phase that constitutes many basic human movements (tasks), typical displacements of joint angles and the state of movement of the center of gravity are empirically determined and stored in the data storage unit 81. Furthermore, multiple assist patterns are assigned to each phase, and different assistance is provided for the same phase using each assist pattern.

[0110] In the above configuration, the motor function improvement device 10 detects changes in the ionic current transmitted from the subject's nervous system to the muscular system using the biosignal detection unit 60 as a biopotential signal that appears on the skin surface, and operates to apply driving force from the drive units (actuators) 82L and 82R based on this detection signal.

[0111] A driving torque corresponding to a biopotential signal generated when the subject voluntarily moves his or her lower limbs is applied as an assist force from the movement mechanism 20 to the subject wearing the movement mechanism 20. In other words, the assist force is a force that generates a torque that acts around each joint (corresponding to the subject's knee joint and hip joint) in the frame mechanism of the movement mechanism 20 as the axis of rotation.

[0112] Therefore, the subject can walk while supporting his or her weight with the combined force of his or her own muscle strength and the drive torque from the drive units 82L and 82R. In addition to walking, the device 10 can also assist the subject in performing movements in accordance with the subject's will, such as when the subject stands up from a seated position in a chair, when the subject sits down in a chair from a standing position, and even when the subject goes up or down stairs. In particular, when a subject has weak muscles, it can be difficult to climb stairs or stand up from a chair. However, when the subject wears the movement mechanism unit 2, the drive torque is applied in accordance with the subject's will, allowing the subject to move without worrying about weakened muscles.

[0113] The synthesis control unit 102 stores the control ratios of the optional control unit 100 and the autonomous control unit 101 set for each phase of each task in the data storage unit 81, and synthesizes the control states of the optional control unit 100 and the autonomous control unit 101 so that the control ratio corresponds to the phase.

[0114] That is, when a subject tries to move his / her body, his / her intention to move is transmitted as a weak ionic current from the brain to the spinal cord, nerves, muscle spindles, and muscles, causing the musculoskeletal system with joints to move. At this time, when a weak bioelectric potential signal is detected from the surface of the subject's skin, the voluntary control unit 100 controls the actuators to move the joints in accordance with the subject's intention.

[0115] Since the movement mechanism 20 is fastened in close contact with the subject's leg (biological part), the driving force of the driving units 82L and 82R is transmitted to the subject as an assisting force for rotating the joint. Therefore, when the subject's body moves due to the assisting force of the movement mechanism 2, signals from the muscle spindles of the Ia afferent neurons return to the brain via the nerves and spinal cord.

[0116] This creates an interactive biofeedback system between the subject, brain, and movement function improving device 10, consisting of a signal transmission system of "brain → spinal cord → motor nerves → [musculoskeletal system + movement mechanism 20]" and a signal transmission system of "movement mechanism 20 → musculoskeletal system (muscle spindles) → sensory nerves → spinal cord → brain." This is bidirectional voluntary control from the brain and the movement mechanism 20, which can further enhance the effects of neurorehabilitation training to restore nervous system function.

[0117] As described above, the motor function improvement device 10 is configured to sense biopotential signals from the brain to the periphery after a decision is made regarding exercise and utilize them for actuator control. It also attempts to capture biopotential signals corresponding to brain activity from peripheral muscle activity. This enables real-time feedback of the peripherally sensed signals to the brain and nervous system. Therefore, by performing rehabilitation with the subject wearing the motor mechanism unit 20, functional recovery in the bidirectional signal transmission system can be promoted.

[0118] In addition, in cases where severe motor dysfunction is present and bioelectric potential signals cannot be detected, voluntary control does not function, so the control ratio is switched so that autonomous control, which controls drive units 82L and 82R using a control program for each phase based on the analysis results of basic human movement patterns and movement mechanisms, can function.

[0119] In this hybrid control method, in which voluntary control and autonomous control coexist, the amplitude and signal characteristics of bioelectrical signals change depending on the state of the body's motor function, even in cases where the body is completely paralyzed or in the progression of an intractable neuromuscular disease, so rehabilitation can be carried out effectively for these conditions as well.

[0120] Furthermore, the synthesis control unit 102 compensates for the physical impedance of the entire system consisting of the entire device and the subject, based on the physical quantities around the joints, in accordance with the physical characteristics of the entire system, including the body characteristics of the subject, and gravity.

[0121] That is, the synthesis control unit 102 constructs a target equation of motion in a calculation environment using the equation of motion data (Mi) and known parameters (Pk) read from the data storage unit 81, and is configured to be able to substitute the drive torque estimate (Te), the joint torque estimate (ΔT), and the joint angle θ into the equation of motion.

[0122] Here, the equation of motion data (Mi) is used to construct the equation of motion for the entire system consisting of the motion function improvement device 10 and the subject, while the known parameters (Pk) consist of dynamic parameters such as the weight of each part of the motion function improvement device 10, the moment of inertia around the joints, the viscosity coefficient, and the Coulomb friction coefficient.

[0123] The joint circumference detection unit 90 includes not only the potentiometer 83, absolute angle sensor 84, and torque sensor 85 described above, but also a relative force detection unit 110, a joint torque estimation unit 111, and a muscle torque estimation unit 112. The relative force detection unit 110 detects the relative force (ΔF) acting on the operation mechanism unit (frame mechanism) 20, that is, a force determined relatively based on the relationship between the force generated by the drive units 82L and 82R and the muscle strength of the subject.

[0124] The joint torque estimating unit 111 estimates the joint moment (ΔT) around each joint of the subject from the difference between the relative force data (ΔF) detected by the relative force detecting unit 110 multiplied by a preset coefficient and the drive torque (Te) detected by the torque sensor 85. The resultant force of the drive torque (Te) of the drive units 82L, 82R and the subject's muscle torque (Tm) acts on the subject's legs as the joint moment (ΔT), so the subject can move his or her legs with less muscle force than if the subject were not wearing the movement mechanism unit (frame mechanism) 20.

[0125] The muscle torque estimator 112 estimates the muscle torque (Tm) due to the subject's muscle force based on the drive torque (Te) detected by the torque sensor 85 and the joint moment (ΔT) estimated by the joint torque estimator 111. Note that the muscle torque (Tm) is calculated to enable parameter identification even when the subject is generating muscle force, and is advantageous when parameter identification is performed while the subject is moving.

[0126] The synthesis control unit 102 performs calculations taking into consideration the drive torque (Te), joint data (θ), joint moment (ΔT), and muscle torque (Tm) obtained from the joint detection unit 90, identifies unknown dynamics parameters (Pu) such as the weight of each part of the subject, the moment of inertia around each joint, viscosity coefficient, Coulomb friction coefficient, etc., and repeats this process multiple times (for example, 10 times) to average them.

[0127] Next, the synthesis control unit 102 reads the ratio (Tm / BES) of the estimated muscle torque (Tm) and the bioelectric potential and a predetermined set gain (Gs) from the data storage unit 81, and if the set gain (Gs) is outside the allowable error range (Ea), it corrects the bioelectric potential (BES) to obtain a corrected bioelectric potential (BES') and makes the ratio (Tm / BES') of the muscle torque (Tm) and the corrected bioelectric potential (BES') approximately equal to the set gain (Gs).

[0128] As a result, it is possible to prevent a situation in which the accuracy of identifying the unknown dynamic parameters (Pu) of the subject is reduced, and also to prevent a situation in which the assisting force generated by the driving units 82L and 82R is too small or too large.

[0129] The synthesis control unit 112 is configured to be able to read the control method data (Ci) from the data storage unit 81, the drive torque (Te), joint torque (ΔT) and joint angle θ obtained from the joint circumference detection unit 90, as well as the identified parameters (Pi) resulting from the identification of the unknown dynamic parameters (Pu), and the corrected bioelectric potential (BES′).

[0130] The synthesis control unit 102 also uses the control method data (Ci) to configure a predetermined control unit in the calculation environment, and is able to send a control signal Ur for controlling the drive of the drive units 82L, 82R by reflecting the drive torque (Te), joint torque (ΔT), joint angle θ, identification parameter (Pi), and bioelectric potential (BES′) in this synthesis control unit 102. The power amplification unit 105 drives the drive units 82L, 82R in accordance with the control signal Ur from the synthesis control unit 102.

[0131] Furthermore, device 10 controls the assist force based on impedance adjustment to eliminate obstacles to natural control caused by constraints due to the physical characteristics of the device itself, i.e., the viscoelasticity around the joints and the inertia of the frame. That is, device 10 calculates joint parameters and compensates for the moment of inertia, viscosity, and elasticity using drive units (actuators) 82L and 82R, thereby improving the assistance rate in walking and reducing discomfort to the subject.

[0132] In this way, the movement function improving device 10 can indirectly change and adjust the characteristics of the subject by changing the characteristics of the entire system, which includes the device itself and the subject. For example, by adjusting the drive torque so as to suppress the effects of the inertial and viscous friction terms of the entire system, it is possible to maximize the subject's inherent ability to perform agile movements, such as reflexes. Furthermore, it is possible to suppress the effects of the subject's own inertial and viscous friction terms, making it possible to have the subject walk faster than their natural cycle or move more smoothly (with less viscous friction) than before wearing the device.

[0133] Furthermore, when the motion function improvement device 10 is attached to the subject, the synthesis control unit 102 identifies the subject's specific dynamics parameters, and the control device 80 controls the drive units 82L and 82R based on the equation of motion into which the identified dynamics parameters are substituted. Therefore, the device can exert an effect according to the control method used by the control device 80, regardless of fluctuation factors such as individual differences and physical condition of the subject.

[0134] Furthermore, since the control device 80 can control the driving units 82L, 82R based on an equation of motion into which the muscle torque (Tm) estimated by the joint circumference detection unit 90 is also substituted, the dynamics parameters can be identified even when the subject is generating muscle force, and the above-mentioned effects can be achieved without requiring the subject to wait for the dynamics parameters to be identified.

[0135] Since the gain between the bioelectric potential (BES) detected by the biosignal detection unit 60 and the muscle torque (Tm) detected by the joint detection unit 90 is adjusted to a preset gain (Gs), it is possible to prevent situations in which the detection results from the biosignal detection unit 60 are insufficiently or excessively sensitive.

[0136] As a result, it is possible to prevent a situation in which the accuracy of identifying the subject's dynamics parameters is reduced, and it is also possible to prevent a situation in which the assisting force generated by the drive units 82L, 82R is too small or too large. Moreover, with the device 10 for improving motor function in this embodiment, calibration can be performed even when the subject is generating muscle force, and the subject does not need to wait for the calibration to be performed.

[0137] At least one of gravity compensation and inertia compensation using the dynamics parameters identified by the synthesis control unit 102 can be applied to the control device 80, thereby preventing a situation in which the weight of the device itself becomes a burden on the subject, or a situation in which the inertia of the device itself causes discomfort to the subject during operation.

[0138] In addition, the synthesis control unit 102 corrects the difference between the subject's movement intention and the motor phenomenon based on a biological self-control loop that is interactively promoted between the subject's body and the movement mechanism unit 20, and at the same time, feedback-adjusts the synthesized control state while correcting the difference using movement commands from the nervous system, so as to minimize the difference between the subject's movement intention and the motor phenomenon.

[0139] As a result, in the motor function improvement device 10, when the subject repeatedly performs voluntary movements using the movement mechanism unit 20, the motor function of the subject's brain, nerves, and muscles can be improved by correcting the difference between the movement commands from the brain and nervous system and the actual motor phenomenon.

[0140] Furthermore, in the motor function improvement device 10, when forming a biological self-control loop, the smallest motor control unit for realizing voluntary movements caused by the subject's will is set as a minimum voluntary movement control unit consisting of the brain nervous system, synaptic connections, and muscular system, and a movement mechanism section 20 is used to establish a biological self-control loop for each minimum voluntary movement control unit that forms linked physical movements to realize a motor phenomenon.

[0141] As a result, in the movement function improvement device 10, by explicitly incorporating the influence of the diseased area and disease cause into the process of function improvement and treatment based on the basic theory of the biological self-control loop by the movement mechanism unit, the influence is extended to other minimum voluntary movement control units, and they are linked in sync with the operation of the movement mechanism unit 20, and the function of each minimum voluntary movement control unit is strengthened and adjusted in sync with the operation of the movement mechanism unit 20 to achieve the target movement, thereby improving the function of the brain, nerves, and muscular system.

[0142] For example, in subjects with progressive diseases (slowly progressive neuromuscular diseases), movement function would normally gradually decline over time, but by using the movement function improvement device 10, a function improvement effect that was previously unthinkable was achieved (Figure 9).Furthermore, while it is commonly believed that muscle destruction in normal life and conventional exercise therapy causes an increase in blood CK levels, which are an indicator of muscle destruction in the blood, the movement function improvement device 10 actually had the effect of reducing CK levels (Figure 10).

[0143] (3) Configuration of a Hippocampal Function Assessment System Using a Movement Function Improvement Device According to the Present Embodiment Figure 11 is a schematic external view of a hippocampal function assessment system 120 according to the present embodiment. This hippocampal function assessment system 120 includes an information acquisition device 130 that is attached to the outer surface of the subject's head to acquire information related to the subject's hippocampal function, an assessment device 140 that evaluates the subject's hippocampal function based on the information acquired by the information acquisition device, and a movement function improvement device 10 that promotes the effects of the subject's physical exercise based on the repetition of a bidirectional biofeedback loop established between the nervous system and the musculoskeletal system.

[0144] The information acquisition device 130 is a device that acquires information related to the hippocampal function of the subject, and in the embodiment shown in Figure 11, for example, an electroencephalogram detection device based on head blood flow measurement described in Japanese Patent Application Publication No. 5717064 and Japanese Patent Application Publication No. 5295584 by the inventor of the present application can be applied.

[0145] In practice, examples of information related to hippocampal function include functional Magnetic Resonance Imaging (fMRI) images, data (images) related to Blood Oxygenation Level Dependent (BOLD) signals, electroencephalogram data (ERP waveforms, time frequencies, etc.), postsynaptic potentials (PSPs), diffusion tensor images, and response data from subjects whose hippocampal function can be evaluated.

[0146] In other words, the information acquisition device 130 of the present invention is not particularly limited in its specific configuration as long as it is capable of acquiring one or more pieces of information related to hippocampal function. Therefore, if the information related to hippocampal function is fMRI images, data (images) related to BOLD signals, and diffusion tensor images, an MRI examination device is used as the information acquisition device 130. If the information related to hippocampal function is EEG data and postsynaptic potentials, an EEG detection device such as a high-density electroencephalograph is used. Furthermore, in the case of a method that exemplifies response data of a subject that can evaluate hippocampal function, a computer device equipped with a user interface (UI) for displaying images and inputting information to the subject is used as the information acquisition device 130.

[0147] The information related to the subject's hippocampal function is information related to hippocampal function acquired during or after a movement assisted by drive in response to bioelectric signals in a subject wearing a movement function improving device 10 that performs drive in response to bioelectric signals generated at the skin surface voluntarily by the subject. This information preferably includes information related to at least one of hippocampal neogenesis, synaptic function, pattern separation, pattern completion, formation of memory traces related to motor procedures in the cerebellum, motor-related functional areas, and somatosensory areas.

[0148] The evaluation device 140 evaluates the hippocampal function of the subject based on the information related to the hippocampal function acquired by the information acquisition device 2.

[0149] The configuration of the evaluation device 140 is not particularly limited, as long as it is capable of evaluating the subject's hippocampal function (synaptic function, pattern separation, pattern completion ability, etc.) in accordance with pre-set evaluation criteria (e.g., threshold values) in response to the form of information related to the subject's hippocampal function acquired by the information acquisition device 130.

[0150] Furthermore, the evaluation device 140 may be configured to evaluate the hippocampal function (improvement of hippocampal function) of the subject by, for example, comparing the data acquired by the information acquisition device 130 with reference data representing hippocampal function. In this case, the reference data may be data regarding hippocampal function previously identified for the subject (or another subject).

[0151] In one embodiment of the hippocampal function evaluation system 120, the evaluation device 140 quantifies the evaluation of the subject's hippocampal function, classifies it into multiple stages according to the numerical value, and can present methods for improving or treating hippocampal function according to each stage.

[0152] 12 is a flowchart of the process of evaluating the hippocampal function of a subject by the hippocampal function evaluation system 1. The process shown in FIG.

[0153] The information acquisition device 130 acquires information related to the subject's hippocampal function (S1). This information is acquired during or after the subject wears the movement function improvement device 10, which performs movements assisted by bioelectric signals generated at the skin surface by the subject's voluntary will. The evaluation device 140 then evaluates the subject's hippocampal function (synaptic function, pattern separation, pattern completion, etc.) based on the information (data) acquired by the information acquisition device 130 (S2).

[0154] In the above configuration, the hippocampal function evaluation system 120 uses the movement function improvement device 10 to implement a repeated two-way biofeedback loop between the cranial nervous system and the musculoskeletal system, which is constructed by the subject's repeated voluntary physical movements while correcting the difference between the movement commands from the cranial nervous system and the actual motor phenomenon, thereby promoting the effect of improving the motor function of the subject's brain, nerves, and muscles.

[0155] In the hippocampal function evaluation system 120, the effects of the subject's physical exercise performed by the motor function improvement device 10 are reflected in this manner, while the information acquisition device 130 acquires information regarding hippocampal function in the subject's brain while dilating peripheral blood vessels, mainly in the subject's head, to improve blood flow and effectively induce synaptic strengthening.

[0156] Next, in the hippocampal function evaluation system 120, if the evaluation device 140 evaluates the hippocampal function, it becomes possible to accurately recognize whether the subject's hippocampal function has been improved.

[0157] The movement function improving device 10 can preferably use HAL (Hybrid Assistive Limb) (registered trademark) manufactured by CYBERDYNE Inc., and examples include HAL Medical Lower Limb Type, HAL Medical Single Joint Type, HAL Independent Living Assistance Lower Limb Type, HAL Independent Living Assistance Single Joint Type, and HAL Lumbar Type (for caregiving / independent living assistance, independent living assistance, and work assistance).

[0158] The lower back-type motion function improving device 10 shown in Figures 4(A) to 5(B) corresponds to the HAL lower back type. HAL is equipped with a control system that has the functional configuration of the basic theory of the voluntary control step, autonomous control step, impedance control step, and bio-autonomic control loop described above.

[0159] The information acquisition method using the information acquisition device 130 of the present invention includes acquiring information related to the hippocampal function of a subject wearing the above-mentioned movement function improving device 10 during or after the subject performs a movement assisted by drive in response to a bioelectric signal. Here, the "movement" is not particularly limited, and examples include leg flexion and extension exercises (squats), elbow flexion and extension exercises, and standing and sitting movements using the movement function improving device 10, and the number of movements can be determined as appropriate.

[0160] The inventors have found that hippocampal function in the subject's brain is improved and hippocampal neogenesis is enhanced by performing the movements (motor training) executed by the movement function improvement device 10 one or more times. Furthermore, the inventors have found that memory traces related to motor procedures are formed in the cerebellum by performing the movements (motor training) executed by the movement function improvement device 10 one or more times.

[0161] That is, in the information acquisition method using the information acquisition device 130 of the present invention, "information related to hippocampal function" includes information on hippocampal neogenesis, synaptic function, pattern separation ability, pattern completion ability, formation of memory traces related to motor procedures in the cerebellum, motor-related functional areas, somatosensory areas, etc. The information acquisition method of the present invention can obtain information that can be used to evaluate the hippocampal function of a subject using a motor function improvement device 10, represented by HAL.

[0162] The information acquisition method of the present invention can also acquire information related to the subject's hippocampal function before the operation by the motor function improvement device 10. Furthermore, in the information acquisition method of the present invention, information related to the subject's hippocampal function acquired during or after the operation by the motor function improvement device 10 can be used to evaluate the effect of improving the subject's hippocampal function, for example, by comparing it with information related to the subject's hippocampal function in the past (before the operation) by the motor function improvement device 10.

[0163] In the information acquisition method of the present invention, as a preliminary step to obtaining information related to the subject's hippocampal function, the subject performs a movement using the motor function improvement device 10. Therefore, the information acquisition method of the present invention can also be considered a low-stress trunk function improvement method (memory trace formation method, synapse strengthening method) and hippocampal function improvement method (cognitive function improvement method, cognitive function maintenance method, cognitive function decline prevention method). Furthermore, while rehabilitation has traditionally been considered to require long-term, repetitive training of limb and trunk function, the information acquisition method of the present invention effectively induces synapse strengthening by using a specific motor function improvement device 10, particularly when combined with imagery training. As a result, when the information acquisition method of the present invention is implemented, not only can information related to the subject's hippocampal function be obtained, but highly effective rehabilitation can be achieved immediately with short training sessions.

[0164] (4) Example 1: Test using HAL Medical Lower Limb Type (4-1) Subjects The subjects (subjects) for the test were 26 healthy volunteers (12 men, 14 women, average age 30±11.2 years) with no particular problems with their walking function.

[0165] (4-2) Method (4-2-1) Gait training using a movement function improvement device A HAL medical lower limb type (manufactured by CYBERDYNE Corporation: Model HAL-ML05) was used as the movement function improvement device 10, and electrodes were attached to the origin of the rectus femoris, vastus lateralis, hamstrings, and gluteus maximus on both sides of the subject to obtain bioelectric potential signals. The torque for assisting hip and knee joint movement was set to level 4.

[0166] The patient underwent approximately 20 minutes of walking training using a weight-relieving walker (A11-in-One, manufactured by ROPOXA / S) to prevent falls during walking. Before and after the 20-minute walking training session while wearing the motor function improvement device 10, functional magnetic resonance imaging (fMRI) was performed on the right knee for a motor task, a right knee for a motor imagery task, resting fMRI, and diffusion tensor imaging. Figure 13 is a schematic diagram showing the data acquisition protocol before and after walking training using the above-mentioned motor function improvement device 10.

[0167] (4-2-2) Data Acquisition Measurement of perceptual changes in lower limb gravity, measurement of quadriceps muscle strength, and measurement of brain activity by fMRI were performed using the following methods with the motor function improvement device 10. Note that, unlike in FIG. 11 , an MRI device was used as the information acquisition device 130.

[0168] To investigate the perceived change in gravity of the lower limbs, the subjects were asked to record on a VAS (visual analog scale) whether their lower limbs felt heavy or light after walking training using the motor function improvement device 10, and the change in the weight of the lower limbs was investigated.

[0169] Quadriceps muscle strength was measured three times in a seated position using a handheld dynamometer μTasF1 (Anima). For activity measurement using fMRI, subjects performed a right knee movement task and a right knee movement imagery task in a block design. Triangular cushions were placed under the subjects' lower limbs to allow knee movement while lying supine on the bed. To prevent motion artifacts due to knee movement, the subjects' thighs and pelvis were secured to the MRI bed with belts.

[0170] The block design consisted of six Rest blocks and five Task blocks, with each block alternately repeated for 15 seconds. Task and Rest instructions were presented on a goggle-type display. Subjects were instructed to remain quiet and not think about anything during the Rest blocks. During the Task block for the right knee movement task, subjects were instructed to repeat knee extension and flexion movements at their own pace. During the right knee movement imagery task, subjects were instructed to imagine knee movement.

[0171] (4-2-3) Acquisition of MRI Images Functional images and structural images were taken using a 3 Tesla MRI (Discovery 750, GE Medicals) as the information acquisition device 130. The imaging sequence was as follows.

[0172] T1-weighted images (T1-WI) (repetition time = 6.9 ms, echo time = 3 ms, flip angle = 15°, matrix = 256 x 256, field-of-view = 256 x 256 mm, thickness = 1 mm, slice gap = 0 mm) and T2*-weighted echo-planar images (repetition time = 3000 ms, echo time = 30 ms, flip angle = 70°, 45 transverse slices, matrix size = 64 x 64, field-of-view = 192 x 192 mm, thickness = 3 mm, slice gap = 0 mm) showing blood oxygen level-dependent contrasts. Functional images were acquired over four sessions, with a total of 220 volumes.

[0173] (4-2-4) Image Analysis SPM12 (https: / / www.fil.ion.ucl.ac.uk / spm / ) was used to analyze fMRI images during the knee movement and knee motor imagery tasks. Image preprocessing included motion correction, slice timing correction, alignment of T1-weighted images with functional images, anatomical standardization of individual brains using an MNI template, and smoothing (6 mm half-width). Functional image analysis involved calculating BOLD contrast images during the motor task and motor imagery task for each individual subject before and after gait training using the motor function improvement device 10. A one-sample t-test was used to identify statistically significant clusters in the 26 subjects for each task (voxel-level threshold of p<0.001 uncorrected, and a cluster-level threshold of FWE (family-wise error)-corrected p<0.05).

[0174] (4-3) Results FIG. 14 shows the change in muscle strength of the quadriceps before and after walking training using the movement function improvement device 10.

[0175] 14, an increase in quadriceps muscle strength was confirmed after walking training using the movement function improvement device 10. Specifically, quadriceps muscle strength before walking training using the movement function improvement device 10 was 5.32±1.25 M / kg, and quadriceps muscle strength after walking training using the movement function improvement device 10 was 6.26±1.46 M / kg, confirming an increase of 19.3±18.0% (mean±SD) before and after walking training using the movement function improvement device 10 (p<0.05, t-test).

[0176] The rate of change in lower limb muscle strength decreased in four cases, but was confirmed to have increased by an average of 19.3±18.0%. Regarding the change in lower limb weight, all cases noticed a lighter feeling in the lower limbs after walking training using the motor function improvement device 10, and a median change of 40% was confirmed.

[0177] Furthermore, before walking training using the motor function improvement device 10, activation of the right anterior cerebellum and activity of the left motor cortex were observed during the right knee movement task, and activation of both posterior cerebellar lobes was observed during the task of imagining right knee movement ( FIG. 15(A) ). On the other hand, after walking training using the motor function improvement device 10, activation of the right anterior cerebellum and increased activity of the left motor cortex were observed during the right knee movement task, and activation of both posterior cerebellar lobes converged to the right during the task of imagining right knee movement ( FIG. 15(A) ).

[0178] Furthermore, walking training using the motor function improvement device 10 efficiently induced activation of sensory information areas that support motor actions, such as the parietal lobe (white arrows on both the right side of the upper row), motor cortex, and allo-somatosensory area of ​​the supplementary motor area (activation occurs on the right side of the lower row, autosomatosensory marking) (Figure 15(B)).

[0179] These results confirm that walking training using the motor function improvement device 10 enhances motor function area activity and promotes the acquisition of motor memory. One of the reasons why such enhancement of the motor function area and the acquisition of memory related to motor procedures are achieved so efficiently is that walking training using the motor function improvement device 10 promotes continuous sensory input, i.e., efficiently carries out sensorimotor transformation.

[0180] Furthermore, simultaneous measurements were taken using fMRI and a 256-channel high-density electroencephalograph before and after gait training using the HAL (both lower limbs type) with the device 10. When the right knee joint was bent and extended, time-frequency analysis revealed a theta wave burst in the right hippocampus (237) after gait training with the device 10 ( FIG. 16 ).

[0181] Furthermore, when a motor (flexion and extension) imagery task was performed, after walking training using the motor function improvement device 10, time-frequency analysis revealed synchronization of theta wave bursts in the right hippocampus (237) and the right posterior cerebellar lobe, Crus I (201) (Figure 17).

[0182] The mechanism behind this effect, which could never be achieved by a single training session performed with standard rehabilitation methods, but was observed after just one training session with the motor function improvement device 10, is presumed to be as follows: It is due to the appearance of activation in the posterior cerebellar lobe Crus I that is synchronized with the hippocampus, and at the same time, the significant activation of the supplementary motor area and M1 may be related to the improvement in quadriceps strength and the reduction in leg weight (subjective lightness).

[0183] Furthermore, when a subject performed a right knee joint flexion and extension imagery task using fMRI, time-frequency analysis confirmed synchronization between the hippocampus and Crus I, the posterior lobe of the cerebellum, after walking training using the motor function improvement device 10, suggesting that memory traces related to motor procedures were formed in the subject's cerebellum through walking training using the motor function improvement device 10. In other words, it is thought that the improvement in motor function due to walking training using the motor function improvement device 10 is due to the formation of memory traces related to motor procedures in Crus I of the cerebellum.

[0184] (5) Example 2: Test using HAL Lumbar Type Independence Support The HAL Lumbar Type Independence Support (manufactured by CYBERDYNE Corporation: Model HAL-FB02) as a motor function improvement device 10 can assist trunk and lower limb exercises for people with weak legs and hips. By wearing it and repeatedly performing trunk movements and standing and sitting movements, it promotes improvement in the function of the body itself, and is expected to increase the level of independence even when HAL is removed. The HAL Lumbar Type Independence Support is easier to wear than the HAL Medical Lower Limb Type (manufactured by CYBERDYNE Corporation: Model HAL-ML05), making training easier even for elderly people.

[0185] (5-1) Subjects The subjects (subjects) for the test were 11 people (5 men and 6 women) aged between 64 and 90 years old with an average age of 80.2±9 years, who were independent in their daily lives, wanted to receive training using the motor function improvement device 10, were capable of performing motor movements of their own volition, and were capable of generating biological signals.

[0186] (5-2) Method (5-2-1) Gait Training Using the Movement Function Improvement Device 10 The movement function improvement device 10 for HAL Lower Back Type independence support has a control system that has the functional configuration of the basic theory of the above-mentioned voluntary control step, autonomous control step, impedance control step, and bio-autonomic control loop, just like HAL Medical Lower Limb Type.

[0187] In this test, the subject wore the HAL Lumbar-Type Movement Function Improvement Device 10 for independent living assistance, and underwent training in a movement mode corresponding to voluntary control steps (CVC: Cybernic Voluntary Control) and training in a movement mode corresponding to autonomous control steps (CAC: Cybernic Autonomous Control).

[0188] Training in a movement mode corresponding to a voluntary control step involves obtaining biopotential signals from the subject's lumbar erector spinae muscles from electrodes attached to the lower back, and the movement function improving device 10 supporting the subject's standing up and other movements based on the biopotential signals. The subject underwent training with support from the movement function improving device 10 for HAL Lumbar Type Independence Assistance.

[0189] The device operates like a robot, using a pre-prepared program based on the analysis of basic human movement patterns and movement mechanisms. In training using a movement mode corresponding to the autonomous control step, electrodes were attached to the subject's lower back, but bioelectric potential signals were not acquired. The subject performed training while receiving movement assistance through the autonomous control of the movement function improvement device 10.

[0190] Both the training group using the movement mode according to voluntary control steps and the training group using the movement mode according to autonomous control steps performed two sets of 10 repetitions of the following exercises once a week: trunk forward / backward bending exercises in a seated position, standing up and sitting down exercises from a chair, and squats. Each training session lasted 20 to 30 minutes.

[0191] (5-2-2) Evaluation method in a hippocampal function evaluation system centered on electroencephalograms The hippocampal function evaluation system 120 used in this test consists of a mobile hippocampal function testing device, and includes an information acquisition unit 130 that acquires electroencephalograms and an evaluation device 140 that evaluates hippocampal function.Furthermore, the information acquisition device 130 is equipped with a touch panel and a trigger box that presents hippocampal memory tasks.

[0192] The information acquisition device 130 can acquire the subject's response data (information related to hippocampal function) to the memory task (correct answer rate, reaction time), and this information is linked to the trigger box, enabling time-frequency analysis along the time axis after the task is administered.

[0193] This hippocampal function evaluation system 120 is configured in consideration of the content of the hippocampal function evaluation method (Patent No. 6328469) developed by the present inventor.

[0194] Specifically, this hippocampal function evaluation system 120 has the subject perform a behavioral analysis task to evaluate the hippocampal pattern separation ability (the ability to recognize similar but slightly different things (Lure), which is carried out by newborn neurons in the dentate gyrus of the hippocampus) and pattern completion ability (the ability to correctly recognize the same thing (Same) as the same, which is carried out by the recurrent laryngeal neural circuit of CA3).The new task (New) can measure the function of the direct pathway of CA1.

[0195] In other words, this hippocampal function evaluation system 120 can easily evaluate the memory circuit network within the hippocampus and measure the state of synaptic function in the direct pathway, trisynaptic pathway, and recurrent pathway.

[0196] This hippocampal function evaluation system 120 can also simultaneously measure postsynaptic potentials (PSPs) from EEG data. Furthermore, this hippocampal function evaluation system 120 can digitize the evaluation results obtained from the test and classify, for example, neurogenesis ability into four levels: normal (I), mild impairment (II), moderate impairment (III), and severe impairment (IV), and can suggest the need for lifestyle improvement for grades I and II, and the need for neuromodulation treatment with drugs or the like for grades III and IV.

[0197] Using this hippocampal function evaluation system 120, a hippocampal function evaluation test (see Patent No. 6,328,469) was conducted to evaluate the hippocampal function of the subject before and after walking training using the motor function improvement device 10.

[0198] (5-3) Results Figure 18 is a diagram showing the percentage of correct answers to the Lure task before and after walking training using the HAL lumbar-type independent living support device 10. As shown in Figure 18, it was confirmed that the percentage of correct answers to the Lure task improved after walking training using the device 10 (post-HAL) compared to before walking training using the device 10 (pre-HAL). In other words, it was confirmed that hippocampal function improved as a result of walking training using the device 10.

[0199] Figure 19 shows the correct answer rate for each task in the hippocampal function evaluation test before walking training using the motor function improvement device 10, and time-frequency analysis. Figure 20 shows the correct answer rate for each task in the hippocampal function evaluation test after walking training using the motor function improvement device 10, and time-frequency analysis.

[0200] [Correction pursuant to Rule 91 02.05.2025] Walking training using the HAL Lumbar Type Movement Function Improvement Device 10 for independent living assistance has produced results similar to those achieved with the HAL Bilateral Lower Limb Type Movement Function Improvement Device 10, with an increase in synaptic function in the direct pathway (EC-CA1; entorhinal cortex to CA1) (left amygdala), CA4 (center arrow), and recurrent pathway (CA3 to CA1) (right arrow) after walking training, with the right anterior (arrow) > left anterior (both reflecting synaptic activity in the right and left hippocampus) electrodes.

[0201] [Correction pursuant to Rule 91 02.05.2025] In a mobile hippocampal function assessment system 120 such as that shown in Figure 11, walking training using a relatively easy-to-wear lumbar HAL-type motor function improvement device 10 for independent living support improved the Lure task success rate (pattern separation ability) from 44% to 56%, and time-frequency analysis showed an increase in the amplitude of theta waves in the trisynaptic pathway in the hippocampal neural network when the Lure task was performed correctly (the arrows in Figure 20 indicate, from left to right, synaptic activity in the CA1 direct pathway, the CA4-CA3 mossy fiber pathway, and the CA3-CAI recurrent pathway to CA1 in the Schaffer pathway). This confirmed that synaptic function had improved (the Lure task success rate, which indicates pattern separation ability, improved from 44% (see Figure 19) to 56% (see Figure 20)).

[0202] (6) Other embodiments In the above-described embodiment, the application of the movement function improvement device 10 equipped with a waist-type movement mechanism unit 20 has been described. However, the present invention is not limited to this, and can be widely applied to movement function improvement devices equipped with a movement mechanism unit adapted to a joint part that allows the subject's body to move, such as a lower body (lower limb type) movement mechanism unit or a single-joint type movement mechanism unit.

[0203] For example, a movement function improving device may be configured that includes a hand-type movement mechanism and a control system that has the functional configuration of the basic theory of the voluntary control step, autonomous control step, impedance control step, and bioautoregulation loop described above. This hand-type movement mechanism may be configured to be directly attached to the subject's fingers or may be configured to be fixed to a table.

[0204] Furthermore, even if the configuration is not as detailed as that of the movement function improvement device 10 in this embodiment, a movement function improvement device with a relatively simple configuration may be applied as long as it has a movement mechanism adapted to the joint parts that allow the subject's body to move.

[0205] For example, a simple motion function improvement device (not shown) may be applied that includes a drive unit that applies power to the subject, a signal detection unit that detects the subject's biopotential signals, a biosignal processing unit that acquires the subject's nerve conduction signals and myoelectric potential signals from the biopotential signals detected by the signal detection unit, an optional control unit that uses the nerve conduction signals and myoelectric potential signals acquired by the biosignal processing unit to generate a command signal for the drive unit to generate power in accordance with the subject's will, and a drive current generation unit that generates a current corresponding to the nerve conduction signal and a current corresponding to the myoelectric potential signal based on the command signal generated by the optional control unit and supplies the current to the drive unit.

[0206] Furthermore, in this embodiment, the evaluation device 140, which evaluates the hippocampal function of the subject in accordance with the information acquired by the information acquisition device 130, has been described as evaluating the hippocampal function of the subject (synaptic function, pattern separation, pattern completion ability, etc.) in accordance with preset evaluation criteria (e.g., threshold values) corresponding to the form of information related to hippocampal function. However, the present invention is not limited to comparison with previously identified reference data representing hippocampal function or multiple-stage classification quantified in accordance with the evaluation of hippocampal function, and may also improve the evaluation accuracy of hippocampal function by utilizing the learning effects of deep learning.

[0207] The evaluation device 140 sequentially analyzes the form of information regarding the subject's hippocampal function acquired by the information acquisition device 130 (hippocampal neogenetic ability, synaptic function, pattern separation ability, pattern completion ability, formation of memory traces related to motor procedures in the cerebellum, motor-related functional areas, somatosensory areas), and extracts feature data corresponding to each of the forms.

[0208] Specifically, the evaluation device 140 extracts partial time series from the information regarding the subject's hippocampal function acquired by the information acquisition device 130 by sequentially dividing the information into predetermined time widths so that there is some overlap in the time direction, and converts each partial time series into data segment images represented in a two-dimensional coordinate system with time and amplitude as the coordinate axes.

[0209] As an example, as shown in Figure 21, the evaluation device 140 uses the sliding window method to extract partial time series from the EEG signals related to the subject's hippocampal function by sequentially dividing them into 2-second intervals with 50% overlap, and then sets each partial time series in a two-dimensional coordinate system that is expanded in the amplitude direction by a factor of 6 to 60, and converts it into a segment image.

[0210] The evaluation device 140 then analyzes the converted segment images in sequence and extracts feature data from each of the segment images.

[0211] Next, the evaluation device 140 uses characteristic patterns classified according to indicators for determining the health of hippocampal function in healthy individuals as training data, and evaluates hippocampal function according to the corresponding indicators from the sequentially extracted characteristic data while referring to a health model constructed by deep learning.

[0212] Specifically, in order to evaluate the hippocampal function of the subject, the evaluation device 140 applies a health model consisting of three modules, namely, a convolutional neural network (CNN) layer, a batch normalization layer (Batch Norm), and an activation function layer (tanh function), as well as a fully connected layer, as shown in FIG. 22 .

[0213] In the health model, fMIR images centered on the hippocampus in the brain from the current frame and the frame two frames before are input, and it is possible to calculate the likelihood for each of several types of hippocampal function (hippocampal neogenesis, synaptic function, pattern separation ability, pattern completion ability, formation of memory traces related to motor procedures in the cerebellum, motor-related functional areas, and somatosensory areas).

[0214] In this way, the evaluation device 140 used a unique dataset to train a health model for each corresponding index through supervised learning using the indices for determining the health of hippocampal function in healthy individuals as training data. For optimization, a cross-entropy loss function and Adam (adaptive movement estimation) with a learning rate of 0.001 were used.

[0215] As a result, when assessing the hippocampal function of a subject, the hippocampal function assessment system 120 can significantly improve the accuracy of assessment of hippocampal function according to the index.

[0216] 10...Movement function improving device, 20...Movement mechanism part, 30...Waist frame, 30A...First waist frame, 30B...Second waist frame, 31...Support, 32, 33...Attachment belt, 40...Left side frame, 41...Right side frame, 42...Battery, 43...Minus button, 44...Plus button, 45...Power button, 50...Thigh fixing part, 50L...Left thigh fixing part, 50R...Right thigh fixing part, 51L, 51R...Stay part, 52L, 52R...Belt part, 60...Biological signal detection output unit, 70...control system, 80...control device, 81...data storage unit, 82L, 82R...drive unit, 83...potentiometer, 84...absolute angle sensor, 85...torque sensor, 90...joint circumference detection unit, 100...optional control unit, 101...autonomous control unit, 102...synthesis control unit, 105...power amplification unit, 110...relative force detection unit, 111...joint torque estimation unit, 112...muscle torque estimation unit, 120...hippocampal function evaluation system, 130...information acquisition device, 140...evaluation device.

Claims

1. A hippocampal function evaluation system comprising: a movement function improvement device for promoting the effect of improving the motor function of a subject's brain-nerve-muscle system by implementing a bidirectional biofeedback loop between the brain-nerve system and the musculoskeletal system, which is constructed while correcting the difference between movement commands from the brain-nerve system and actual motor phenomena through the repetition of the subject's voluntary physical movements; an information acquisition device for acquiring information related to the subject's hippocampal function during or after the operation of the movement function improvement device while the movement function improvement device is worn by the subject; and an evaluation device for evaluating the hippocampal function of the subject based on the information acquired by the information acquisition device.

2. The hippocampal function evaluation system according to claim 1, characterized in that the information includes information on at least one of hippocampal neogenetic ability, synaptic function, pattern separation ability, pattern completion ability, formation of memory traces related to motor procedures in the cerebellum, motor-related functional areas, and somatosensory areas.

3. The hippocampal function evaluation system described in claim 1 or 2, characterized in that the evaluation device sequentially analyzes the form of the information acquired by the information acquisition device, extracts feature data corresponding to each of the forms, and then evaluates hippocampal function according to the corresponding indicator from the sequentially extracted feature data while referring to a health model constructed by deep learning, using feature patterns classified according to indicators for determining the health of hippocampal function in healthy individuals as training data.

4. A hippocampal function evaluation system as claimed in claim 1 or 2, characterized in that the subject's voluntary physical movement is performed using a movement function improvement device comprising: a drive unit that applies power to the subject; a signal detection unit that detects the subject's biopotential signals; a biosignal processing unit that acquires the subject's neurotransmitter signals and myoelectric potential signals from the biopotential signals detected by the signal detection unit; a voluntary control unit that uses the neurotransmitter signals and myoelectric potential signals acquired by the biosignal processing unit to generate a command signal for the drive unit to generate power in accordance with the subject's will; and a drive current generation unit that generates a current corresponding to the neurotransmitter signals and a current corresponding to the myoelectric potential signals based on the command signal generated by the voluntary control unit and supplies the current to the drive unit.

5. A movement mechanism unit that is used to be integrated with the subject and has a drive unit that is driven actively or passively in conjunction with the subject's physical movement; a signal detection unit that detects changes in ionic current transmitted from the subject's nervous system to the subject's muscular system as bioelectric potential signals that appear on the skin surface; a joint circumference detection unit that detects physical quantities around the joints that accompany the subject's physical movement based on the output signal from the drive unit; a voluntary control unit that controls the drive unit to produce a movement phenomenon that reflects the subject's intention to move based on the bioelectric potential signals and the physical quantities around the joints; and an autonomous control unit that stores reference parameters for each phase, which is a series of smallest movement units that make up the subject's movement pattern classified as a task, in a data storage unit, compares the physical quantities around the joints with the reference parameters stored in the data storage unit, estimates the phase of the subject's task, and controls the drive unit to generate power according to the phase.

3. The hippocampal function evaluation system of claim 1 or 2, characterized in that the subject's voluntary physical movement is performed using a movement function improvement device comprising: a data storage unit that stores the control ratios of the voluntary control unit and the autonomous control unit set for each phase of each task; and a synthesis control unit that synthesizes the control states of the voluntary control unit and the autonomous control unit so as to achieve a control ratio corresponding to the phase; and the synthesis control unit compensates the physical impedance of the entire system consisting of the entire device and the subject to match the physical characteristics of the entire system including the subject's body characteristics and gravity based on the physical quantities around the joints, and feedback-adjusts the synthesized control state while correcting the difference by movement commands from the nervous system based on a biological autocontrol loop that is interactively promoted between the subject's body and the movement mechanism unit so as to minimize the difference between the subject's movement intention and the motor phenomenon.

6. The hippocampal function evaluation system described in claim 5, characterized in that when forming the biological self-control loop, the smallest motor control unit for realizing voluntary movements caused by the subject's will is set as a smallest voluntary movement control unit consisting of the nervous system, synaptic connections, and muscular system, and the biological self-control loop is established for each smallest voluntary movement control unit that forms linked physical movements to realize the motor phenomenon using the operating mechanism part.

Images